Was anybody at the Sean Olive talk at CanJam?

[deadkrillin]

I was there and attended the talk. There will be a YouTube video of it soon. He showed listening results that the original 2019 Harnan target translated to the 5128 was rates the highest compard IE targets proposed by Headphones.com and the BK5128 DF.

Wow! Thanks so much for the reply Dr. Olive. I’ll keep my eyes out for the video.

The in-room flat curve I measured with 19 speakers in our Reference Room is identical to the 5128 DF.

Oh wow, so you expanded the amount of HATS measurements from the old -30 degree/0 degree/30 degree arrangement from the prior studies! That’s super interesting.

I’ll hold any questions until after the video is posted. Thank you again for popping in!

 

[Sean Olive]

Here are the 6 targets that were tested. All measured on a 5128 with the listening test design described below.

 

[Sean Olive]

Listening test design here:

 

[Sean Olive]

Wow! Thanks so much for the reply Dr. Olive. I’ll keep my eyes out for the video.

Oh wow, so you expanded the amount of HATS measurements from the old -30 degree/0 degree/30 degree arrangement from the prior studies! That’s super interesting.

I’ll hold any questions until after the video is posted. Thank you again for popping in!

I measured 19 different channels in our Reference Room with each speaker equalized to be flat steady-state at the listening position. I looked at different combinations from front channels, surrounds, heights and all combinations.

 

[Sean Olive]

Just a few comments because I was not there.

This is very unlikely to be an adequate target because DF is an unnatural and largely impossible acoustic condition. Olive made roughly the same comments in his papers about why the Harman headphone target became necessary.


Somewhat more probable, but in the end the targets are not dissimmilar from each other and there's not much that can be expected from headphones that have not been individually EQed using in-ear measurements. We know LF region will differ in tilt, curve and boost by a few dB, and on a person susceptible to leakage effects and preference. MF will be flattish on a fixture and on a person. HF will be elevated due to ear resonance but totally dependent on the dimensions of the ear and the final tuning based again on preference within a few dB.

Harman already did the baseline research about subjective preference and basic requirements (such as preventing high-Q resonances), and produced a defensible target. Beyond introducing a requirement for individual EQ and changes to source material standards for music presented through headphones, I don't know what other progress we can expect. Changing the shape of the curve by a few dB is not important. What would be much more valuable is if Harman produced recommendations about new headphone designs that show better FR stability across many people.

Just to be clear. The BK5128 DF curve unmodified was the least preferred headphone target curve tested. This confirms previous listening tests where the DF wasincluded. It is too bright and thin, yet it remains the current standard and recommended target quoted in AES and IEC standards for the past 30 years.

The in-room measurements of the 19 loudspeakers is close to the DF target because the large number of sources coming from all directions approximate a DF and the speakers are equalized from anechoic flat to in-room flat meaning the speakers will also sound thin and bright.

If you apply a bass shelving and treble filter to the headphone matched to the flat in-room loudspeakers you can approximate the predicted-in room response of an anechoically flat loudspeaker. The downward slope of that curve is somewhere between 7-10 dB. There is a tolerance window around that target that will satisfy most people's taste. Some of the variance in taste is related to age (hearing loss), listening experience and possibly gender.

The HARMAN 2019 IE Target has a bass shelf boost and a treble shelf cut so is not DF tuned. The new HARMAN target we included in the test was the BK5128 DF with a bass and treble adjustment used in the HARMAN 2018 AE target: 6.6 dB bass shelf and -1.4 dB treble cut -- so it was not DF. I called it HARMAN Beta 2024 because it has not gone through any method of adjustment experiments where listeners modify the bass/treble to according to taste.

 

[Thomas_A]

Is there a binaural target curve published?

 

[deadkrillin]

Here are the 6 targets that were tested. All measured on a 5128 with the listening test design described below.

View attachment 356188

These all look very similar except for differences in the ~6kHz region: this makes me wonder if future testing specifically around the variation in preference in this region is a potentially good idea.

I measured 19 different channels in our Reference Room with each speaker equalized to be flat steady-state at the listening position. I looked at different combinations from front channels, surrounds, heights and all combinations.

Right, it makes sense that it would get closer to a Diffuse Field the more measurements you take at different angles and heights. Thanks again for your replies!

 

[VietnamRadioEnjoyer]

Is there a binaural target curve published?

based on this paper by Meta, you need a flat FR (at eardrum) for binaural. so your best bet is an IEM EQ'd flat (after bass region)

 

[IAtaman]

based on this paper by Meta, you need a flat FR (at eardrum) for binaural. so your best bet is an IEM EQ'd flat (after bass region)

Is that really what the paper is saying?

Results show that the audio content had a significant effect on listeners’ preference of target HpTF. For spatial audio content, individual flat was the preferred choice, which agrees with previous research [1 , 2 , 3]. For stereo content, on the other hand, Harman obtained higher ratings. This confirms the hypothesis that the audio content type should be taken into account when HpEQ is considered, if the aim is to optimize listener preference and individual HpEQ is available. For generic HpEQ, on the other hand, this effect was less evident, as generic flat and Harman targets did not show significant differences for spatial audio content.

Bold is added by me.

What it I understand is that it says is if you have the possibility to generate individual EQ for yourself, that is probably your best bet, and if you don't, and are going to use a generic EQ, Harman target is as good as any.

 

[Thomas_A]

My experience is that bass is boosted too high using the Harman when listening to binaural recordings.

 

[VietnamRadioEnjoyer]

Is that really what the paper is saying?

Results show that the audio content had a significant effect on listeners’ preference of target HpTF. For spatial audio content, individual flat was the preferred choice, which agrees with previous research [1 , 2 , 3]. For stereo content, on the other hand, Harman obtained higher ratings. This confirms the hypothesis that the audio content type should be taken into account when HpEQ is considered, if the aim is to optimize listener preference and individual HpEQ is available. For generic HpEQ, on the other hand, this effect was less evident, as generic flat and Harman targets did not show significant differences for spatial audio content.

Bold is added by me.

What it I understand is that it says is if you have the possibility to generate individual EQ for yourself, that is probably your best bet, and if you don't, and are going to use a generic EQ, Harman target is as good as any.

they used an off-ear headphone in the test which still interacts with the pinna so to get individual flat you need an in-ear mic. that is why I recommended IEMs

the way I understand it is that harman adds the ear-gain because listening to stereo speakers our ears expect that. but with binaural the ear-gain is baked into the recording itself. Flat FR for binaural is also what oratory1990 suggests & in my (biased) experience for binaural I preferred flat FR+IEMs

also take look at binaural recording devices (this one is neumann KU 100) and how mics are placed inside the ear-canal

 

[Sean Olive]

Was anybody at the Sean Olive talk at CanJam & if so, do you have pictures of his slides?

A friend who attended said Sean revealed he's not using "Harman In-Room Flat" anymore & is using the same 5128 DF baseline as Crinacle & Headphones.com, which was suspicious and unexpected.

He said Sean also did small listening tests that showed that trained listeners tended to prefer Harman's new beta target, Harman IE 2019, SoundGuys' target, and Headphones.com's target roughly equally... which is also suspicious and unexpected. Hoping someone here has some pictures because I really don't know what to think.

Was anybody at the Sean Olive talk at CanJam & if so, do you have pictures of his slides?

A friend who attended said Sean revealed he's not using "Harman In-Room Flat" anymore & is using the same 5128 DF baseline as Crinacle & Headphones.com, which was suspicious and unexpected.

He said Sean also did small listening tests that showed that trained listeners tended to prefer Harman's new beta target, Harman IE 2019, SoundGuys' target, and Headphones.com's target roughly equally... which is also suspicious and unexpected. Hoping someone here has some pictures because I really don't know what to think.

Here is the difference between the BK5128 DF and the average response of 19 loudspeakers equalized to be flat measured with the BK 5128. They are almost identical. So I think it is valid to use it as a starting point to make bass and treble adjustments. The shape of the curve is not so different from the flat-in room response measured in the previous study using a different loudspeaker but same room.

 

[Thomas_A]

they used an off-ear headphone in the test which still interacts with the pinna so to get individual flat you need an in-ear mic. that is why I recommended IEMs

the way I understand it is that harman adds the ear-gain because listening to stereo speakers our ears expect that. but with binaural the ear-gain is baked into the recording itself. Flat FR for binaural is also what oratory1990 suggests & in my (biased) experience for binaural I preferred flat FR+IEMs

also take look at binaural recording devices (this one is neumann KU 100) and how mics are placed inside the ear-canal

Shouldn't bass be baked in as well? Listening to binaurally recorded natural sounds such as a basketball bouncing around you, you can hear that bass is boosted compared to a natural bouncing basketball, if you use a "Harman bass boosted"-headphone. If flat, it sounds much more like the real thing.

 

[Sean Olive]

based on this paper by Meta, you need a flat FR (at eardrum) for binaural. so your best bet is an IEM EQ'd flat (after bass region)

they used an off-ear headphone in the test which still interacts with the pinna so to get individual flat you need an in-ear mic. that is why I recommended IEMs

the way I understand it is that harman adds the ear-gain because listening to stereo speakers our ears expect that. but with binaural the ear-gain is baked into the recording itself. Flat FR for binaural is also what oratory1990 suggests & in my (biased) experience for binaural I preferred flat FR+IEMs

also take look at binaural recording devices (this one is neumann KU 100) and how mics are placed inside the ear-canal

Yes, if the ear canal resonance is in the recording you need to take it out of the headphone on playback. Make the headphone flat.

If you make blocked ear recordings then you need to include the ear canal during playback: The trick is how to identify the contribution of the ear canal and the blocked external ear when calibrating the headphones.

You can do this by dividing the responses at DRP of a test mannequin versus the blocked canal. Easy to do with a test mannequin. Harder to do for a calibration of an individual listener canal unless you use a probe mic or can estimate their response at DRP with some tricky science.

 

[Sean Olive]

Shouldn't bass be baked in as well? Listening to binaurally recorded natural sounds such as a basketball bouncing around you, you can hear that bass is boosted compared to a natural bouncing basketball, if you use a "Harman bass boosted"-headphone. If flat, it sounds much more like the real thing.

See my comment about lack of tactile information in binaural reproduction. It can account for up to 3 db of bass that isn’t perceived in the binaural versus the in situ experience. We published some AES papers on this a few years ago based on experiments in a car.

 

[Thomas_A]

See my comment about lack of tactile information in binaural reproduction. It can account for up to 3 db of bass that isn’t perceived in the binaural versus the in situ experience. We published some AES papers on this a few years ago based on experiments in a car.

Yet I feel it gets too much bass. If I record something over speakers (in ear mics) with natural room gain included, it gets too much bass when played over headphones with Harman boosted bass. Have to reduce bass to get balance.

 

[Sean Olive]

I'm saying 3db above flat for binaural. Harman Target is 6.4 dB bass shelf and the IE target is 10 dB on a type 3.3 ear simulator

 

[Four Kneez]

The talk has been posted on YouTube.

 

[markanini]

Which IE target was mislabeled? Was the mislabeled target using BK5128DF+2015 Harman filters, the result of subtracting Harman OE2015 with Harman in-room?

 

[ObjectiveSubjectivist]

The talk has been posted on YouTube.

Thanks will watch it later